It is proposed to examine the appropriateness of traditional survey methods for the study of older respondents. The major sources of response error and sample bias such as non-response, interviewer effects, question format, mode of administration, and response set are addressed and aspects which may create particular difficulty for aged respondents are described. Analyses are outlined to establish the significance of these various sources of error for older respondents. These analyses will focus on older respondents, but other age groups will also be examined in order to determine whether any observed effects are exclusively a phenomenon of older age or are not age related. Relevant data will be drawn from existing surveys, thereby providing for highly cost-effective research procedures. A monograph describing the results of these investigations including conclusions and recommendations based upon the results is planned and will serve as reference for researchers designing survey instruments for use with aged populations and for anybody using published reports of research data on older people.